CN111784381A - Privacy protection and SOM network-based power customer segmentation method and system - Google Patents

Privacy protection and SOM network-based power customer segmentation method and system Download PDF

Info

Publication number
CN111784381A
CN111784381A CN202010520053.1A CN202010520053A CN111784381A CN 111784381 A CN111784381 A CN 111784381A CN 202010520053 A CN202010520053 A CN 202010520053A CN 111784381 A CN111784381 A CN 111784381A
Authority
CN
China
Prior art keywords
power
neural network
som
clustering
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010520053.1A
Other languages
Chinese (zh)
Other versions
CN111784381B (en
Inventor
杨杨
喻鹏
孙寅栋
严泽凡
张振威
颜拥
姚影
王刘旺
王健鑫
刘祖龙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing University of Posts and Telecommunications
Electric Power Research Institute of State Grid Zhejiang Electric Power Co Ltd
Original Assignee
Beijing University of Posts and Telecommunications
Electric Power Research Institute of State Grid Zhejiang Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing University of Posts and Telecommunications, Electric Power Research Institute of State Grid Zhejiang Electric Power Co Ltd filed Critical Beijing University of Posts and Telecommunications
Priority to CN202010520053.1A priority Critical patent/CN111784381B/en
Publication of CN111784381A publication Critical patent/CN111784381A/en
Application granted granted Critical
Publication of CN111784381B publication Critical patent/CN111784381B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/231—Hierarchical techniques, i.e. dividing or merging pattern sets so as to obtain a dendrogram
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/602—Providing cryptographic facilities or services
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
    • G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
    • G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

The embodiment of the invention provides a power customer segmentation method and a system based on privacy protection and an SOM network, wherein the method comprises the following steps: acquiring power utilization multidimensional data of a power customer through an intelligent ammeter; clustering the electricity utilization multidimensional data to obtain the clustering number and the initial clustering center of the power customer; inputting the electricity utilization multidimensional data into an improved SOM neural network, and taking the cluster number and the initial cluster center as initial values of the improved SOM neural network to obtain a power customer segmentation result; wherein, the improved SOM neural network is obtained by improving the SOM neural network through an equal deviation theory. The embodiment of the invention improves the accuracy of the SOM network in the subdivision of the power network customers, can realize the rapid and effective clustering of the customers when the number of the customers is larger, and greatly reduces the blindness and subjectivity of manually appointing the clustering number.

Description

Privacy protection and SOM network-based power customer segmentation method and system
Technical Field
The invention relates to the technical field of power systems, in particular to a power customer segmentation method and system based on privacy protection and an SOM network.
Background
With the development of smart grids and the deep advancement of structural reform of the power industry, power customers play an increasingly important role in the power market, and therefore, the subdivision of the power customers needs to be strengthened, so that different sales and service strategies are adopted for different power customers. The main idea of subdividing the customers is to utilize various existing effective methods to collect, classify and analyze the customers and the requirements thereof, and then classify and manage the risks and the values of customer groups with different behavior characteristics, so that the service level of enterprises is improved, and personalized service is realized. In the electric power market, the customers are subdivided, so that power supply enterprises can know the electricity utilization habits of the customers and identify the value customers, and personalized service strategies and differentiated marketing strategies can be made, so that the service level is improved.
In recent years, many domestic colleges and power research institutes have started to study the electricity consumption customer segmentation in the power system, and many electricity consumption customer segmentation strategies have been proposed. Many scholars mainly develop value-based power customer segmentation research from the perspective of power supply enterprises. In the aspect of segmentation technology, a clustering method which is simple to operate is mostly adopted, but the situation of a large number of clients is not considered in much research. With the increase of the number of customers, it is time-consuming to compute the value of the customers one by one and then re-cluster, and the increase of noise and isolated point data directly influences the clustering effect, so that the existing power customer segmentation effect is poor.
Therefore, there is a need for a power customer segmentation method and system based on privacy protection and SOM network to solve the above problems.
Disclosure of Invention
Aiming at the problems in the prior art, the embodiment of the invention provides a power customer segmentation method and system based on privacy protection and an SOM network.
In a first aspect, an embodiment of the present invention provides a power customer segmentation method based on privacy protection and an SOM network, including:
acquiring power utilization multidimensional data of a power customer through an intelligent ammeter;
clustering the electricity utilization multidimensional data to obtain the clustering number and the initial clustering center of the power customer;
inputting the electricity utilization multidimensional data into an improved SOM neural network, and taking the cluster number and the initial cluster center as initial values of the improved SOM neural network to obtain a power customer segmentation result; wherein, the improved SOM neural network is obtained by improving the SOM neural network through an equal deviation theory.
Further, the clustering the electricity consumption multidimensional data to obtain the clustering number and the initial clustering center of the electricity customers includes:
clustering the power customers by a hierarchical clustering method of UMPGA, constructing a power customer list, and obtaining the clustering number corresponding to the information field of each power customer;
obtaining the score of each node in the power customer list, taking the node with the highest score as a dense area of the power customer, and adding the dense area into a linked list;
and classifying the power customers according to the scores of all the nodes, and acquiring initial clustering centers of the power customers according to classification results.
Further, before the inputting the electricity utilization multidimensional data into the improved SOM neural network and taking the cluster number and the initial cluster center as initial values of the improved SOM neural network to obtain a power customer segmentation result, the method further includes:
s1, acquiring K data points with the highest scores in the sample power customer list;
s2, initializing the connection weight, the learning rate, the neighborhood radius and the internal deviation of the clustering unit of the SOM neural network through the K data points to obtain an initialized SOM neural network;
s3, based on the initialized SOM neural network, determining winning neurons by using electricity multi-dimensional sample data, competition layer neurons and internal deviation of clustering units;
s4, updating the connection weight of the winning neuron and updating the connection weight of the adjacent neuron to obtain an updated SOM neural network;
s5, adjusting the learning rate and the adjacent radius of the updated SOM neural network to obtain an adjusted SOM neural network;
s6, obtaining the deviation value of the whole input space in the adjusted SOM neural network based on the equal deviation theory, and updating the connection weight of the winning neuron and the connection weight of the adjacent neuron;
and S7, repeating the steps S4 to S6 until preset conditions are met, and obtaining an improved SOM neural network to perform power customer segmentation on the power customers.
Further, the obtaining a deviation value of the whole input space in the adjusted SOM neural network based on the equal deviation theory includes:
dividing the input space S of the adjusted SOM neural network into K small spaces { S1，S2，...，SKAnd acquiring a competition layer neuron Z of the adjusted SOM neural network according to the divided K small spacesiThe formula is as follows:
Figure BDA0002531706630000031
wherein xl∈SiAnd is and
Figure BDA0002531706630000032
wherein S isiDenotes the ith cell, xiRepresenting input samples in the ith small space, xlRepresentation space SiThe ith sample point in (1);
from input samples x of each small spaceiAnd competition layer neuron ZiObtaining the clustering unit internal deviation D (S) of each small spacei) The formula is as follows:
Figure BDA0002531706630000033
according to the clustering unit internal deviation D (S) of each small spacei) And acquiring the deviation value of the whole input space in the adjusted SOM neural network.
Further, the acquiring, by the smart meter, the electricity consumption multidimensional data of the electricity consumer includes:
acquiring power consumption information data of a power customer through an intelligent ammeter;
and preprocessing the electricity information data and the electricity data indexes to obtain electricity multidimensional data of the electricity customers, wherein the preprocessing comprises the elimination of missing values and abnormal values and normalization processing.
Further, the power data metrics include: average electricity price index, current electricity consumption, customer arrearage rate index, customer credit, customer electricity consumption increase rate, valley electricity consumption rate, electricity consumption increase contribution rate, historical contemporaneous electricity consumption increase rate, current electricity consumption recovery rate, and historical contemporaneous electricity consumption increase rate.
Further, the acquiring, by the smart meter, the electricity consumption information data of the electricity consumer includes:
acquiring power consumption information encryption data of a power customer, wherein the power consumption information encryption data is obtained by encrypting the power consumption information data of the power customer through an RSA encryption algorithm by an intelligent electric meter;
and decrypting the electricity utilization information encrypted data to obtain user information data of the power customer.
In a second aspect, an embodiment of the present invention provides a power customer segmentation system based on privacy protection and SOM network, including:
the power consumption data acquisition module is used for acquiring power consumption multidimensional data of power customers through the intelligent ammeter;
the power utilization data clustering module is used for clustering the power utilization multidimensional data to obtain the clustering number and the initial clustering center of the power customers;
the electric power customer segmentation module is used for inputting the electricity utilization multidimensional data into an improved SOM neural network, and taking the cluster number and the initial cluster center as initial values of the improved SOM neural network to obtain an electric power customer segmentation result; wherein, the improved SOM neural network is obtained by improving the SOM neural network through an equal deviation theory.
In a third aspect, an embodiment of the present invention provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the method provided in the first aspect when executing the program.
In a fourth aspect, an embodiment of the present invention provides a non-transitory computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the method as provided in the first aspect.
According to the power customer segmentation method and system based on privacy protection and SOM, provided by the embodiment of the invention, the network connection weight is initialized by a method of detecting the center of a data-dense area, so that the connection weight close to each category center is initialized, and the convergence probability and the learning speed can be improved; meanwhile, the deviation is adjusted to guide the learning of the SOM neural network, the accuracy of the SOM network on the subdivision of the power network customers is improved, when the number of the customers is large, the customers can be clustered quickly and effectively, and the blindness and subjectivity of manually specifying the clustering number are greatly reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 is a schematic flowchart of a power customer segmentation method based on privacy protection and SOM network according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a portion of sample data after preprocessing according to an embodiment of the present invention;
FIG. 3 is a diagram illustrating an initialization result of a part of weights according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating updated weights according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a power customer segmentation system based on privacy protection and SOM network according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
With the development of smart grids and the advance of power reform businesses, the quality of service of power customers is more and more prominent. Marketing strategies for small groups of similar customers are more efficient and of higher quality than marketing strategies employed by large and broad-faced groups of customers. The subdivision of the electricity consumption customers is beneficial to making an accurate and effective marketing strategy and improving the service quality of the electricity consumption customers. The current power customers have extremely large data volume and a certain number of isolated points, so that great challenges are brought to the subdivision of the power customers. Meanwhile, in a power consumption information acquisition system of a smart grid, hundreds of millions of smart meters need to continuously acquire power consumption information data of users from a home terminal and transmit the data to a control center through a building domain network and a regional network. If data is leaked in the processes of storage, processing, transmission and the like, the electricity utilization rule of the user can be exposed, and privacy information such as living habits of the user can be deduced from the electricity utilization rule, so that the property and personal safety of the user are threatened. Therefore, in the process of collecting information, it is particularly important to protect the privacy of the user when collecting information, and if the privacy of the user cannot be well protected, the power data of the user is highly likely to be utilized maliciously.
The embodiment of the invention introduces a dense initialization theory, initializes the network connection weight by detecting the center of a data dense area, so that the connection weight close to the center of each category can be initialized to improve the convergence probability and the learning speed.
Fig. 1 is a schematic flowchart of a power customer segmentation method based on privacy protection and an SOM network according to an embodiment of the present invention, and as shown in fig. 1, an embodiment of the present invention provides a power customer segmentation method based on privacy protection and an SOM network, including:
step 101, acquiring power utilization multidimensional data of a power customer through an intelligent ammeter;
in the embodiment of the present invention, the power control center acquires the electricity consumption multidimensional data of the power customer through the encrypted transmission data sent by the smart meter, and preferably, in the embodiment of the present invention, the acquired electricity consumption multidimensional data is subjected to data preprocessing, so that subsequent subdivision of the power customer is more accurate.
And 102, clustering the electricity utilization multidimensional data to obtain the clustering number and the initial clustering center of the power customer.
In the embodiment of the present invention, after a set of electricity consumption multidimensional data is acquired, the set of electricity consumption multidimensional data is clustered by using a non-weighted clustering algorithm (UMPGA) of arithmetic mean, so as to obtain a cluster number and an initial cluster center of the set of electricity consumption multidimensional data.
Step 103, inputting the electricity utilization multidimensional data into an improved SOM neural network, and taking the cluster number and the initial cluster center as initial values of the improved SOM neural network to obtain a power customer segmentation result; wherein, the improved SOM neural network is obtained by improving the SOM neural network through an equal deviation theory.
According to the electric power customer subdivision method based on privacy protection and the SOM, provided by the embodiment of the invention, the network connection weight is initialized by detecting the center of the data dense area, so that the convergence probability and the learning speed can be improved by initializing the connection weight close to each category center; meanwhile, the deviation is adjusted to guide the learning of the SOM neural network, the accuracy of the SOM network on the subdivision of the power network customers is improved, when the number of the customers is large, the customers can be clustered quickly and effectively, and the blindness and subjectivity of manually specifying the clustering number are greatly reduced.
On the basis of the above embodiment, the clustering the electricity consumption multidimensional data to obtain the cluster number and the initial cluster center of the power customer includes:
clustering the power customers by a hierarchical clustering method of UMPGA, constructing a power customer list, and obtaining the clustering number corresponding to the information field of each power customer;
obtaining the score of each node in the power customer list, taking the node with the highest score as a dense area of the power customer, and adding the dense area into a linked list;
in the embodiment of the invention, the hierarchical clustering algorithm based on UMPGA firstly clusters the power customers through the electricity multidimensional data, thereby constructing a power customer list and forming a cluster number in the information field of each power customer; then, the node with the highest score is obtained through the algorithm, the node is a dense area of the power customer, the node is added to the linked list, and in the dense area of the power customer, the center of the category is likely to exist. In the embodiment of the present invention, the node score is the average similarity × the number of power customers.
And classifying the power customers according to the scores of all the nodes, and acquiring initial clustering centers of the power customers according to classification results.
Further, according to the node scores obtained in the above embodiments, the power customers in the power customer list are classified, and the average weight of the customer data in the power customer dense area is taken according to the weight of each dimension of the center vector, so as to be used for selecting the centers of the power customer dense areas, that is, obtaining the initial clustering center of the power customers.
On the basis of the above embodiment, before the inputting the electricity utilization multidimensional data into the improved SOM neural network and taking the cluster number and the initial cluster center as initial values of the improved SOM neural network to obtain a power customer segmentation result, the method further includes:
step S1, K data points with the highest score in the sample power customer list are obtained;
and step S2, initializing the connection weight, the learning rate, the neighborhood radius and the internal deviation of the clustering unit of the SOM neural network through the K data points to obtain the initialized SOM neural network.
In the present inventionIn the embodiment of the invention, the cluster number and the initial cluster center of the sample power client list are obtained from the K data points with the highest score in the sample power client list, so as to initialize the connection weight of the SOM neural network according to the cluster number and the initial cluster center, initialize the connection weight w, the learning rate ao, the neighborhood radius Nbo and the cluster unit internal deviation D (S)i)＝1，i＝1，2，…，K。
Step S3, based on the initialized SOM neural network, through the electricity multi-dimensional sample data XjCompetitive layer neurons ZiAnd clustering unit internal deviation D (S)i) Determining a winning neuron, wherein the formula is as follows:
Figure BDA0002531706630000081
wherein, ∑-1(Xj,Zi) Represents XjAnd ZiThe covariance matrix of (2).
Step S4, updating the connection weight of the winning neuron and the connection weight of the adjacent neuron to obtain an updated SOM neural network;
and step S5, adjusting the learning rate and the adjacent radius of the updated SOM neural network to obtain the adjusted SOM neural network.
In the embodiment of the invention, as the number of iterations increases, the learning rate and the adjacent radius decrease, and in order to ensure the convergence of the algorithm, the learning speed and the neighborhood radius need to be adjusted, so that only the winning neuron is adjusted in learning.
And step S6, acquiring the deviation value of the whole input space in the adjusted SOM neural network based on the equal deviation theory, and updating the connection weight of the winning neuron and the connection weight of the adjacent neuron.
In the embodiment of the present invention, the deviation value D (S) of each of the competition layer neuron internal classes is calculated based on the equal deviation theoryi) I.e. the intra-clustering unit bias. Aiming at the difference of the winning times of the neurons in the competition layer, different formulas are adopted to obtain D (S)i) In particular, if obtainedThe winning number is not 0, and the formula is as follows:
Figure BDA0002531706630000082
where num represents the number of neuron wins, wiRepresents the connection weight, x, of the ith contention layer neuronjDenotes the jth input sample, non _ zero denotes the number of neurons with a non-zero number of wins ∑-1(xj,wi) Denotes xjAnd wiA covariance matrix;
if the winning number is 0, the formula is:
Figure BDA0002531706630000083
where zero denotes the number of neurons with a winning number of zero.
Further, the bias value D (S) for each competition layer neuron internal class obtained according to the above embodimenti) And acquiring the deviation value of the whole input space in the adjusted SOM neural network.
And S7, repeating the steps S4 to S6 until preset conditions are met, and obtaining an improved SOM neural network to perform power customer segmentation on the power customers.
In the embodiment of the invention, the improved SOM neural network is obtained by continuously iterating until the maximum iteration number is met or the algorithm converges, so that the power customers are subdivided according to the improved SOM neural network.
On the basis of the above embodiment, the obtaining a deviation value of the whole input space in the adjusted SOM neural network based on the equal deviation theory includes:
dividing the input space S of the adjusted SOM neural network into K small spaces { S1，S2，...，SKAnd acquiring a competition layer neuron Z of the adjusted SOM neural network according to the divided K small spacesiThe formula is as follows:
Figure BDA0002531706630000091
wherein xl∈SiAnd is and
Figure BDA0002531706630000092
wherein S isiDenotes the ith cell, xiRepresenting input samples in the ith small space, xlRepresentation space SiThe ith sample point in (1); competition layer neurons ZiIs a small space SiIn the present embodiment, is the center of (A) with the closest SiIs expressed by the arithmetic center point of (1);
from input samples x of each small spaceiAnd competition layer neuron ZiObtaining the clustering unit internal deviation D (S) of each small spacei) The formula is as follows:
Figure BDA0002531706630000093
in the embodiment of the invention, D (S) is passedi) Indicating a deviation of class i, i.e. from the small space SiTo a small space SiThe sum of the distances of the centers.
According to the clustering unit internal deviation D (S) of each small spacei) And acquiring the deviation value of the whole input space in the adjusted SOM neural network.
In the embodiment of the invention, based on the equal deviation theory, the deviation value D (S) of the whole input space in the adjusted SOM neural network is minimum, and each small space S is obtained by calculation in the embodimentiD (S) ofi) Calculating the whole deviation D (S) of the input space S in the SOM neural network, wherein the formula is as follows:
Figure BDA0002531706630000101
on the basis of the above embodiment, the acquiring, by the smart meter, the electricity consumption multidimensional data of the electricity consumer includes:
acquiring power consumption information data of a power customer through an intelligent ammeter;
and preprocessing the electricity information data and the electricity data indexes to obtain electricity multidimensional data of the electricity customers, wherein the preprocessing comprises the elimination of missing values and abnormal values and normalization processing.
On the basis of the above embodiment, the power data index includes: average electricity price index, current electricity consumption, customer arrearage rate index, customer credit, customer electricity consumption increase rate, valley electricity consumption rate, electricity consumption increase contribution rate, historical contemporaneous electricity consumption increase rate, current electricity consumption recovery rate, and historical contemporaneous electricity consumption increase rate.
On the basis of the above embodiment, the acquiring, by the smart meter, the electricity consumption information data of the electricity consumer includes:
acquiring power consumption information encryption data of a power customer, wherein the power consumption information encryption data is obtained by encrypting the power consumption information data of the power customer through an RSA encryption algorithm by an intelligent electric meter;
and decrypting the electricity utilization information encrypted data to obtain user information data of the power customer.
In the embodiment of the invention, the intelligent electric meter transmits the electricity utilization information data of the electric power customer to the control center through the RSA encryption algorithm, and the control center acquires the electricity utilization data of the electric power customer after decryption; then, the electricity consumption data of the power customer is preprocessed together with indexes such as average electricity price index, current electricity consumption, customer defaulting rate index, customer credit degree, customer electricity consumption increasing rate, valley electricity consumption rate, electricity quantity increasing contribution rate, electricity charge increasing contribution rate, historical contemporaneous electricity charge increasing rate, current electricity charge recovery rate, historical contemporaneous electricity charge increasing rate and the like, so that electricity consumption multidimensional data of the power customer is obtained.
According to the embodiment of the invention, the privacy of the power customers is greatly protected by using the RSA encryption algorithm to transmit data, and meanwhile, the accuracy of subsequent power customer subdivision is higher due to the preprocessed power utilization multidimensional data.
In an embodiment of the present invention, K is set to 5, that is, power customers are classified into five types: excellent customers, high-quality customers, stable customers, potential arrearage risk customers and potential attrition risk customers; in the examples of the present invention, 0 to 4 represent, in order: excellent customers, high-quality customers, stable customers, potential arrearage risk customers and potential attrition risk customers; further, 18 samples are selected from a data set to be used for analysis, and the samples in the data set are multidimensional data indexes of the power customers in recent years, and comprise indexes such as current electricity consumption, average electricity price index, customer defaulting rate index, customer electricity utilization increase rate, customer credit degree, low-valley electricity utilization rate, electricity quantity increase contribution rate, electricity charge increase contribution rate, historical contemporaneous electricity charge increase rate, current-phase electricity charge recovery rate and historical contemporaneous electricity quantity increase rate.
Further, the data obtained in the above embodiment is preprocessed, including missing value and outlier elimination, and data normalization processing, fig. 2 is a schematic diagram of a part of the preprocessed sample data provided in the embodiment of the present invention, and the preprocessed data can be referred to as fig. 2. Then, clustering the power customers according to the preprocessed data, in the embodiment of the present invention, classifying the power customer list according to the obtained scores by using a hierarchical clustering method of UMPGA; the classification results were as follows: [ 041203013401242311 ], wherein 0 to 4 correspond in sequence to: excellent customers, premium customers, stable customers, potential arrearage risk customers, and potential churn risk customers.
Further, initializing a connection weight of the SOM neural network by using K data points, and initializing the connection weight, the learning rate, the neighborhood radius and the internal deviation of a clustering unit. The initial learning rate is 0.05, and the neighborhood radius is 5. Fig. 3 is a schematic diagram of an initialization result of a part of weights according to an embodiment of the present invention.
On the basis of the above embodiment, it is next necessary to update the connection weights of the winning neurons, and at the same time, update the connection weights of the adjacent neurons, and fig. 4 is a schematic diagram of the updated weights according to the embodiment of the present invention. The connection weight of the winning neuron and the connection weight of the adjacent neuron are iteratively updated until the maximum iteration times or algorithm convergence is met, finally, the power customer subdivision result is obtained, and the final classification result is as follows: [ 011203103404212312 ], wherein 0 to 4 correspond in sequence to: excellent customers, premium customers, stable customers, potential arrearage risk customers, and potential churn risk customers, i.e., the original specimen classification results are: [ excellent customer, premium customer, stable customer, excellent customer, potential arrearage risk customer, premium customer, excellent customer, potential arrearage risk customer, potential attrition risk customer, excellent customer, potential attrition risk customer, stable customer, excellent customer, potential arrearage risk customer, premium customer, stable customer ].
Fig. 5 is a schematic structural diagram of an electric power customer segmentation system based on privacy protection and an SOM network according to an embodiment of the present invention, and as shown in fig. 5, an electric power customer segmentation system based on privacy protection and an SOM network according to an embodiment of the present invention includes an electricity consumption data obtaining module 501, an electricity consumption data clustering module 502, and an electric power customer segmentation module 503, where the electricity consumption data obtaining module 501 is configured to obtain electricity consumption multidimensional data of an electric power customer through a smart meter; the electricity utilization data clustering module 502 is used for clustering the electricity utilization multidimensional data to obtain the clustering number and the initial clustering center of the power customer; the electric power customer segmentation module 503 is configured to input the electricity utilization multidimensional data into the improved SOM neural network, and use the cluster number and the initial cluster center as initial values of the improved SOM neural network to obtain an electric power customer segmentation result; wherein, the improved SOM neural network is obtained by improving the SOM neural network through an equal deviation theory.
According to the power customer subdivision system based on privacy protection and the SOM, provided by the embodiment of the invention, the network connection weight is initialized by a method for detecting the center of the data dense area, so that the connection weight close to each category center is initialized, and the convergence probability and the learning speed can be improved; meanwhile, the deviation is adjusted to guide the learning of the SOM neural network, the accuracy of the SOM network on the subdivision of the power network customers is improved, when the number of the customers is large, the customers can be clustered quickly and effectively, and the blindness and subjectivity of manually specifying the clustering number are greatly reduced.
The system provided by the embodiment of the present invention is used for executing the above method embodiments, and for details of the process and the details, reference is made to the above embodiments, which are not described herein again.
Fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, and referring to fig. 6, the electronic device may include: a processor (processor)601, a communication Interface (Communications Interface)602, a memory (memory)603 and a communication bus 604, wherein the processor 601, the communication Interface 602 and the memory 603 complete communication with each other through the communication bus 604. The processor 601 may call logic instructions in the memory 603 to perform the following method: acquiring power utilization multidimensional data of a power customer through an intelligent ammeter; clustering the electricity utilization multidimensional data to obtain the clustering number and the initial clustering center of the power customer; inputting the electricity utilization multidimensional data into an improved SOM neural network, and taking the cluster number and the initial cluster center as initial values of the improved SOM neural network to obtain a power customer segmentation result; wherein, the improved SOM neural network is obtained by improving the SOM neural network through an equal deviation theory.
In addition, the logic instructions in the memory 603 may be implemented in the form of software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program is implemented by a processor to execute the privacy protection and SOM network-based power customer segmentation method provided in the foregoing embodiments, for example, including: acquiring power utilization multidimensional data of a power customer through an intelligent ammeter; clustering the electricity utilization multidimensional data to obtain the clustering number and the initial clustering center of the power customer; inputting the electricity utilization multidimensional data into an improved SOM neural network, and taking the cluster number and the initial cluster center as initial values of the improved SOM neural network to obtain a power customer segmentation result; wherein, the improved SOM neural network is obtained by improving the SOM neural network through an equal deviation theory.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (10)

1. A power customer segmentation method based on privacy protection and SOM network is characterized by comprising the following steps:
acquiring power utilization multidimensional data of a power customer through an intelligent ammeter;
clustering the electricity utilization multidimensional data to obtain the clustering number and the initial clustering center of the power customer;
inputting the electricity utilization multidimensional data into an improved SOM neural network, and taking the cluster number and the initial cluster center as initial values of the improved SOM neural network to obtain a power customer segmentation result; wherein, the improved SOM neural network is obtained by improving the SOM neural network through an equal deviation theory.
2. The privacy protection and SOM network-based power customer segmentation method according to claim 1, wherein the clustering process is performed on the power utilization multidimensional data to obtain the clustering number and the initial clustering center of the power customer, and comprises the following steps:
clustering the power customers by a hierarchical clustering method of UMPGA, constructing a power customer list, and obtaining the clustering number corresponding to the information field of each power customer;
obtaining the score of each node in the power customer list, taking the node with the highest score as a dense area of the power customer, and adding the dense area into a linked list;
and classifying the power customers according to the scores of all the nodes, and acquiring initial clustering centers of the power customers according to classification results.
3. The power customer segmentation method based on privacy protection and SOM network as claimed in claim 2, wherein before inputting the electricity multidimensional data into the modified SOM neural network and using the cluster number and the initial cluster center as initial values of the modified SOM neural network to obtain the power customer segmentation result, the method further comprises:
s1, acquiring K data points with the highest scores in the sample power customer list;
s2, initializing the connection weight, the learning rate, the neighborhood radius and the internal deviation of the clustering unit of the SOM neural network through the K data points to obtain an initialized SOM neural network;
s3, based on the initialized SOM neural network, determining winning neurons by using electricity multi-dimensional sample data, competition layer neurons and internal deviation of clustering units;
s4, updating the connection weight of the winning neuron and updating the connection weight of the adjacent neuron to obtain an updated SOM neural network;
s5, adjusting the learning rate and the adjacent radius of the updated SOM neural network to obtain an adjusted SOM neural network;
s6, obtaining the deviation value of the whole input space in the adjusted SOM neural network based on the equal deviation theory, and updating the connection weight of the winning neuron and the connection weight of the adjacent neuron;
and S7, repeating the steps S4 to S6 until preset conditions are met, and obtaining an improved SOM neural network to perform power customer segmentation on the power customers.
4. The power customer segmentation method based on privacy protection and SOM network as claimed in claim 3, wherein the obtaining the deviation value of the whole input space in the adjusted SOM neural network based on the equal deviation theory comprises:
dividing the input space S of the adjusted SOM neural network into K small spaces { S1，S2，...，SKAnd acquiring a competition layer of the adjusted SOM neural network according to the divided K small spacesNeuron ZiThe formula is as follows:
Figure FDA0002531706620000021
wherein xl∈SiAnd is and
Figure FDA0002531706620000023
wherein S isiDenotes the ith cell, xiRepresenting input samples in the ith small space, xlRepresentation space SiThe ith sample point in (1);
from input samples x of each small spaceiAnd competition layer neuron ZiObtaining the clustering unit internal deviation D (S) of each small spacei) The formula is as follows:
Figure FDA0002531706620000022
according to the clustering unit internal deviation D (S) of each small spacei) And acquiring the deviation value of the whole input space in the adjusted SOM neural network.
5. The privacy protection and SOM network-based power customer segmentation method according to claim 1, wherein the obtaining of the power customer's power consumption multidimensional data through a smart meter comprises:
acquiring power consumption information data of a power customer through an intelligent ammeter;
and preprocessing the electricity information data and the electricity data indexes to obtain electricity multidimensional data of the electricity customers, wherein the preprocessing comprises the elimination of missing values and abnormal values and normalization processing.
6. The privacy protection and SOM network-based power customer segmentation method of claim 5, wherein the power data metrics comprise: average electricity price index, current electricity consumption, customer arrearage rate index, customer credit, customer electricity consumption increase rate, valley electricity consumption rate, electricity consumption increase contribution rate, historical contemporaneous electricity consumption increase rate, current electricity consumption recovery rate, and historical contemporaneous electricity consumption increase rate.
7. The privacy protection and SOM network-based power customer segmentation method according to claim 5, wherein the step of obtaining the power utilization information data of the power customer through a smart meter comprises the following steps:
acquiring power consumption information encryption data of a power customer, wherein the power consumption information encryption data is obtained by encrypting the power consumption information data of the power customer through an RSA encryption algorithm by an intelligent electric meter;
and decrypting the electricity utilization information encrypted data to obtain user information data of the power customer.
8. A power customer segmentation system based on privacy protection and SOM network, comprising:
the power consumption data acquisition module is used for acquiring power consumption multidimensional data of power customers through the intelligent ammeter;
the power utilization data clustering module is used for clustering the power utilization multidimensional data to obtain the clustering number and the initial clustering center of the power customers;
the electric power customer segmentation module is used for inputting the electricity utilization multidimensional data into an improved SOM neural network, and taking the cluster number and the initial cluster center as initial values of the improved SOM neural network to obtain an electric power customer segmentation result; wherein, the improved SOM neural network is obtained by improving the SOM neural network through an equal deviation theory.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program performs the steps of the privacy preserving and SOM network based power consumer segmentation method according to any one of claims 1 to 7.
10. A non-transitory computer readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the power customer segmentation method based on privacy protection and SOM network according to any one of claims 1 to 7.
CN202010520053.1A 2020-06-09 2020-06-09 Power customer subdivision method and system based on privacy protection and SOM network Active CN111784381B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010520053.1A CN111784381B (en) 2020-06-09 2020-06-09 Power customer subdivision method and system based on privacy protection and SOM network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010520053.1A CN111784381B (en) 2020-06-09 2020-06-09 Power customer subdivision method and system based on privacy protection and SOM network

Publications (2)

Publication Number Publication Date
CN111784381A true CN111784381A (en) 2020-10-16
CN111784381B CN111784381B (en) 2023-05-16

Family

ID=72755870

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010520053.1A Active CN111784381B (en) 2020-06-09 2020-06-09 Power customer subdivision method and system based on privacy protection and SOM network

Country Status (1)

Country Link
CN (1) CN111784381B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115829418A (en) * 2023-02-07 2023-03-21 国网江苏省电力有限公司营销服务中心 Power consumer load characteristic portrait construction method and system suitable for load management

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106296315A (en) * 2016-11-03 2017-01-04 广东电网有限责任公司佛山供电局 Context aware systems based on user power utilization data
CN108280479A (en) * 2018-01-25 2018-07-13 重庆大学 A kind of power grid user sorting technique based on Load characteristics index weighted cluster algorithm
US20190369963A1 (en) * 2016-12-05 2019-12-05 Hitachi, Ltd. Data processing system and data processing method
CN110851502A (en) * 2019-11-19 2020-02-28 国网吉林省电力有限公司 Load characteristic scene classification method based on data mining technology

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106296315A (en) * 2016-11-03 2017-01-04 广东电网有限责任公司佛山供电局 Context aware systems based on user power utilization data
US20190369963A1 (en) * 2016-12-05 2019-12-05 Hitachi, Ltd. Data processing system and data processing method
CN108280479A (en) * 2018-01-25 2018-07-13 重庆大学 A kind of power grid user sorting technique based on Load characteristics index weighted cluster algorithm
CN110851502A (en) * 2019-11-19 2020-02-28 国网吉林省电力有限公司 Load characteristic scene classification method based on data mining technology

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115829418A (en) * 2023-02-07 2023-03-21 国网江苏省电力有限公司营销服务中心 Power consumer load characteristic portrait construction method and system suitable for load management

Also Published As

Publication number Publication date
CN111784381B (en) 2023-05-16

Similar Documents

Publication Publication Date Title
US20220335501A1 (en) Item recommendations using convolutions on weighted graphs
CN108898154A (en) A kind of electric load SOM-FCM Hierarchical clustering methods
CN111898839B (en) Importance degree classification method and device for power users
CN113839926B (en) Method, system and device for modeling intrusion detection system based on characteristic selection of wolf algorithm
CN113111924A (en) Electric power customer classification method and device
CN111062428A (en) Hyperspectral image clustering method, system and equipment
Hewapathirana et al. Change detection in noisy dynamic networks: a spectral embedding approach
CN111415167B (en) Network fraud transaction detection method and device, computer storage medium and terminal
CN110348516B (en) Data processing method, data processing device, storage medium and electronic equipment
Cheng et al. An unsupervised ensemble framework for node anomaly behavior detection in social network
Chen et al. Learning category distance metric for data clustering
CN111784381B (en) Power customer subdivision method and system based on privacy protection and SOM network
CN113688906A (en) Customer segmentation method and system based on quantum K-means algorithm
Diao et al. Clustering by detecting density peaks and assigning points by similarity-first search based on weighted K-nearest neighbors graph
CN114186168A (en) Correlation analysis method and device for intelligent city network resources
Minoungou et al. Learning an MR-sort model from data with latent criteria preference direction
Wang et al. Multi-view group anomaly detection
CN104636489B (en) The treating method and apparatus of attribute data is described
CN114238852A (en) Operation data analysis method and device, storage medium and electronic equipment
CN114219522A (en) Customer consumption behavior prediction method and device, electronic equipment and storage medium
CN113706173A (en) Information management method and device, electronic equipment and storage medium
CN114443409A (en) Payment business system monitoring method, device and equipment and computer storage medium
CN111709478A (en) Fuzzy clustering method and device based on anchor graph
CN115905887B (en) Big data intelligent decision system based on data processing
CN115563555B (en) Data-driven Internet card user value classification method, device and medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant